Heat shock proteins (hsp's) protect cells from a variety of stresses. Human breast cancer cells may express high levels of hsp27 and hsp70 in particular, which we have found to be associated with general tumor aggressiveness. Preliminary evidence also suggests that hsp's play a role in drug resistance, and understanding the mechanisms involved could lead to clinical strategies to circumvent such resistance and improve patient survival. We initially found that heat shock increases the resistance of breast cancer cells to doxorubicin, while inducing hsp27 and hsp70. Introducing hsp27 cDNA makes the cells resistant to doxorubicin, while blocking hsp27 expression with flavones (e.g. quercitin) reverses resistance. We now need to determine whether hsp70 plays a similar role. We also plan to further investigate means of modulating hsp27 expression for therapeutic benefit by manipulating its regulatory promoter system - we have already identified key elements of the promoter region to be targeted, along with a novel DNA-binding protein which binds one of these elements. We will examine mechanisms which may be involved in the association of hsp's with drug resistance, and confirm the association in clinical breast cancer specimens from doxorubicin-resistant vs. naive patients. Our Specific Aims are: (1) To confirm our preliminary finding that hsp70 may also play a role in doxorubicin resistance in human breast cancer cells. We will use antisense oligonucleotides to inhibit expression of both hsp70 and its constitutive cognate hsc70 in breast cancer cells, and full-length cDNAs to induce overexpression, determining drug resistance in soft agar cloning assays and in vivo nude mouse studies. (2) To further develop pharmacologic means (flavone inhibition) and molecular means (promoter studies) to circumvent hsp27-induced doxorubicin resistance. (3) To search for mechanisms associated with this resistance, focusing in particular on the role of topoisomerase II. (4) To translate these findings to the clinical setting by determining the relationship of hsp27 and hsp70 with clinical doxorubicin resistance. We will compare hsp levels in a set of 100 doxorubicin-resistant metastases vs. 100 naive breast cancer specimens, and will also correlate hsp levels with clinical outcome in a prospective, randomized adjuvant clinical trial (SWOG 8897) involving doxorubicin. All of this will prepare for a Phase I clinical trial directed at circumventing hsp-induced doxorubicin resistance using the pharmacologic agent quercitin, though the trial itself is not a part of the present proposal, and will also suggest other approaches for reversing resistance to this otherwise most useful drug in breast cancer treatment.